Skip to main content
BenchLM
Advertise

Model comparison

Qwen 3.6 Max (preview) vs Step 3.7 Flash

Data verified

Head-to-head evidence from 15 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.

59.44/100
Margin
8.7pts
← winning
50.76/100
0 category wins2 category wins

BenchAlign evidence: Qwen 3.6 Max (preview) supported; Step 3.7 Flash estimated. Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.

Evidence parity. Qwen 3.6 Max (preview) and Step 3.7 Flash share 15 comparable benchmark results. 2 of 8 categories are comparable. 7 results are unique to Qwen 3.6 Max (preview); 15 to Step 3.7 Flash.

Updated July 16, 2026
Shared results
15
Qwen 3.6 Max (preview) only
7
Step 3.7 Flash only
15
Comparable categories
2 / 8

Pick Qwen 3.6 Max (preview) if you want the stronger benchmark profile. Step 3.7 Flash only becomes the better choice if coding is the priority.

Confidence note. This is a partial-evidence comparison with 15 shared benchmark results across 5 evidence categories; 2 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.

Why this result

Qwen 3.6 Max (preview) is clearly ahead on the provisional aggregate, 68 to 57. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

Category breakdown

Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.

Category scores and score margins for Qwen 3.6 Max (preview) and Step 3.7 Flash
CategoryQwen 3.6 Max (preview)ΔStep 3.7 Flash
CodingQwen 3.6 Max (preview)51.0Margin→ 5.3Step 3.7 Flash56.3
AgenticQwen 3.6 Max (preview)65.4Margin→ 1.0Step 3.7 Flash66.4
KnowledgeQwen 3.6 Max (preview)73.9MarginNo overlapStep 3.7 FlashNot measured
MathQwen 3.6 Max (preview)18.4MarginNo overlapStep 3.7 FlashNot measured

Decisive benchmark drivers

The largest measured benchmark gaps in this matchup, with exact reported values.

More
A · Qwen 3.6 Max (preview)B · Step 3.7 Flash

  1. Terminal-Bench 2.0

    Agentic
    Source ↗
    A 65.4%B 59.5%
    Winner: Qwen 3.6 Max (preview)Δ 5.9
    Terminal-Bench 2.0: Qwen 3.6 Max (preview) scored 65.4%; Step 3.7 Flash scored 59.5%. Qwen 3.6 Max (preview) wins this benchmark.

  2. SWE-bench Pro

    Coding
    Source ↗
    A 57.3%B 56.3%
    Winner: Qwen 3.6 Max (preview)Δ 1
    SWE-bench Pro: Qwen 3.6 Max (preview) scored 57.3%; Step 3.7 Flash scored 56.3%. Qwen 3.6 Max (preview) wins this benchmark.

Operational comparison

Runtime and commercial metrics are compared only when both models have a complete sourced value.

MetricQwen 3.6 Max (preview)Step 3.7 FlashComparison
Input / output priceUSD per 1M tokensQwen 3.6 Max (preview)Not availableStep 3.7 Flash$0.2 input / $1.15 outputA complete price comparison is not available.
Generation speedtokens per secondQwen 3.6 Max (preview)Not availableStep 3.7 FlashNot availableA complete speed comparison is not available.
First-answer latencyseconds to first tokenQwen 3.6 Max (preview)Not availableStep 3.7 FlashNot availableA complete latency comparison is not available.
Context windowmaximum listed tokensQwen 3.6 Max (preview)256KStep 3.7 Flash256KListed context windows are equal.

Benchmark Deep Dive

AgenticStep 3.7 Flash wins
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
Terminal-Bench 2.0Source 65.4%59.5%Qwen 3.6 Max (preview) leads
QwenClawBenchSource 59.0%—Not comparable
QwenWebBenchSource 1532—Not comparable
τ²-bench resultsSource 95.9%98.5%Step 3.7 Flash leads
BrowseCompSource —75.8%Not comparable
DeepSearchQASource —92.8%Not comparable
GDPval-AASource —25.9%Not comparable
ToolathlonSource —49.5%Not comparable
Claw-EvalSource —67.1%Not comparable
HLE w/ toolsSource —47.2%Not comparable
Gert LabsSource —51.57%Not comparable
AA Agentic IndexSource —21.5%Not comparable
GDPval-AASource —1017Not comparable
APEX-Agents-AASource —14.8%Not comparable
CodingStep 3.7 Flash wins
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
SWE-bench ProSource 57.3%56.3%Qwen 3.6 Max (preview) leads
SciCodeSource 47%—Not comparable
NL2RepoSource 42.9%—Not comparable
Terminal-Bench 2.0Source 65.4%59.5%Qwen 3.6 Max (preview) leads
Terminal-Bench HardSource 43.9%35.6%Qwen 3.6 Max (preview) leads
AA-SciCodeSource 46.9%40.0%Qwen 3.6 Max (preview) leads
AA Coding IndexSource —39.6%Not comparable
Reasoning
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
AA-LCRSource 69.7%63.7%Qwen 3.6 Max (preview) leads
CritPtSource 3.7%2.3%Qwen 3.6 Max (preview) leads
Knowledge
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
SuperGPQASource 73.9%—Not comparable
Artificial Analysis Intelligence IndexSource 40.0%30.3%Qwen 3.6 Max (preview) leads
AA-GPQA DiamondSource 88.8%80.9%Qwen 3.6 Max (preview) leads
AA-HLESource 28.9%19.9%Qwen 3.6 Max (preview) leads
AA-Omniscience IndexSource 10.2%-37.5%Qwen 3.6 Max (preview) leads
AA-Omniscience AccuracySource 37.7%25.4%Qwen 3.6 Max (preview) leads
AA-Omniscience Hallucination RateSource 44.2%84.4%Qwen 3.6 Max (preview) leads
Math
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
FrontierMath v2 (Tiers 1-3)Source 23.103%—Not comparable
FrontierMath v2 (Tier 4)Source 4.167%—Not comparable
Multimodal
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
SimpleVQASource —79.2%Not comparable
V*Source —95.3%Not comparable
AA-MMMU-ProSource —75.3%Not comparable
Design Arena WebsiteSource —1218Not comparable
Inst. Following
BenchmarkQwen 3.6 Max (preview)Step 3.7 FlashResult
AA-IFBenchSource 76.6%67.3%Qwen 3.6 Max (preview) leads
Frequently Asked Questions (3)

Which is better, Qwen 3.6 Max (preview) or Step 3.7 Flash?

Qwen 3.6 Max (preview) is ahead on BenchLM's provisional leaderboard, 68 to 57. The biggest single separator in this matchup is Terminal-Bench 2.0, where the scores are 65.4% and 59.5%.

Which is better for coding, Qwen 3.6 Max (preview) or Step 3.7 Flash?

Step 3.7 Flash has the edge for coding in this comparison, averaging 56.3 versus 51. Inside this category, Terminal-Bench Hard is the benchmark that creates the most daylight between them.

Which is better for agentic tasks, Qwen 3.6 Max (preview) or Step 3.7 Flash?

Step 3.7 Flash has the edge for agentic tasks in this comparison, averaging 66.4 versus 65.4. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.

Related Comparisons

Explore More

Alibaba Compare Pricing Methodology Find Your Best LLM Overall Rankings
Last updated: July 16, 2026

The AI models change fast. We track them for you.

A weekly brief for engineers and researchers covering new models, ranking shifts, and pricing changes.

Free. No spam. Unsubscribe anytime.